Has anyone looked at Earl Geddes Nathan10?

[ThomasW]

Just to crank up the pressure you might post even more of the pictures along with Earl's refusal in the DIYaudio Nathan thread...

I almost bought a pair of these, I'd be livid if that's what I received..

[A9X]

Just to crank up the pressure you might post even more of the pictures along with Earl's refusal in the DIYaudio Nathan thread...

I almost bought a pair of these, I'd be livid if that's what I received..

I was doing that when I noticed this post.

Livid! That hardly begins to decribe how I feel.

It has been made all the harder to take as not an hour before I picked up these from the PO, I collected 8 AE drivers (4 x TD15X, 4 x TD10M), which look absolutely superb, even better than I had hoped.

[rc white]

Just as I have long suspected given the various tolerances involved and the dimensional stability of the materials used, you are better off making a device using a simple circular arc joining the throat to a conical horn and thence a circular mouth radius blending to the baffle.
rcw

[Saurav]

Wow. I'm sorry, this is just wrong. I hope he ends up fixing this.

[JoshK]

I can understand the dissapointment. I was a bit upset myself when I saw that the support was done out of particle board. My plastic mounts are out of round a bit, much more than .001". Still, mine are fairly useable compared to what A9X got.

[Dennis H]

Just to crank up the pressure you might post even more of the pictures along with Earl's refusal in the DIYaudio Nathan thread...

I almost bought a pair of these, I'd be livid if that's what I received..

Oh my again at the quality of the woodwork in the knockdown kit. And he defends it saying people are just being picky! Unmuckingbelievable.

[Sakura]

If the interface from the tweeter's exit to the edge of the waveguid was all polished smooth (without the awful uneven thick edges in AW9's pictures) and so precise as Dr. Geddes is claiming it needs to be (which, I don't believe is true) - THEN, then, I would say: $50 per waveguide would be a FAIR price for the 10" one.

IP value? Yes, some, that's why I'd pay $50 instead of $15.

Maybe if I HEARD the speakers, if they really sound good? Well then, I guess $200 per waveguide (plus lots of work making it look ok and finishing) might be tolerable. But having not heard them, and seeing the sloppy workmaship, wow, that looks bad. Particularly after hearing Dr. Geddes go on and on about how hard it is to make them, and how much of an investment he's had to put in to "get it right".

I'm sure if somebody makes available a CAD file with the proper profile, some of us DIY folks could turn out a few pairs of WGs ourselves that are better built with less slop.

[JoshK]

We already know the profile, but I don't know CAD.

[Dennis H]

Just as I have long suspected given the various tolerances involved and the dimensional stability of the materials used, you are better off making a device using a simple circular arc joining the throat to a conical horn and thence a circular mouth radius blending to the baffle.
rcw

Yup, I agree. Doing some quick calcs -- if you were stacking 3/4" wood to make the horn, a 1" roundover bit would make a pretty good throat to transition to a 45 degree degree chamfer bit to make the body of the horn. You want the roundover to be hitting 45 degrees at the top of the first piece. A 1.5" roundover bit would transition the mouth to the baffle. I think you could certainly get within the 1mm tolerance that Geddes now claims he's shooting for when it comes to making a real horn and maybe get a bit closer with some sanding in the throat.

That's just doing it with hand tools. I don't have any experience with Solidworks but if we could get the oblate spheroid equation in there, it would be easy to CNC the slices -- assuming we know exactly how thick the wood is as the circles' radius grows by the wood thickness. Maybe it's not worth it to CNC as a 15" horn would only be 10 slices of 3/4" so it wouldn't be a big deal to do it by hand.

[mikela]

This looks like an interesting approach:

Image not available

[Deward Hastings]

Oh my again at the quality of the woodwork in the knockdown kit. And he defends it saying people are just being picky!

To Geddes' defence (sort of) . . . he obviously came to this project from an academic background, with little experience in manufacturing and even less in social psychology or managing customer expectations. I think he honestly did not grasp that representations of outstanding sonic quality would be read as representing outstanding build quality as well, or that his customers would expect build quality at least as good as, for example, the enclosures included in "Madisound" or "Parts Express" kits. He would have been *far* better off if he'd just sold the waveguide as "fresh from the mold", with all the trimming and filling and sprue removal to be done by the builder (as it is with model airplane kits) and sold the enclosures only as plans, or completed (and built by a shop that knows how). How Linkwitz markets ORION and PLUTO might have been a good example to study.

At this point it's entirely on him to acknowledge his error, apologize to his customers, and get on with doing what he should have done in the first place. Insulting his customers for their expectations, whether he "encouraged" those expectations or not, is most definitely *not* the way to go . . .

[JoshK]

...and get on with doing what he should have done in the first place. Insulting his customers for their expectations . . .

That is the way I read it at first, which is half true.

[Dennis H]

To Geddes' defence (sort of) . . . he obviously came to this project from an academic background, with little experience in manufacturing and even less in social psychology or managing customer expectation

Well, it's a bit more than just managing expectations. The guy flat can't run a router. A 3rd grader could do better. And cutting a side panel 1/8" short and saying you should just fill the gap with putty.... c'mon, gimme a break. And saying there's no way to flush the woofer with the baffle because of the thickness of the wood -- like no one has ever done a rabbet cut before.

The good news is John J. is going to be cutting his wood from now on.

Oh, and while I'm in rant mode, he was talking about how he might invent a way to extract the center channel from the LR channels -- apparently he's never heard of DPL2. And he goes on saying it can't be done in the analog domain but he could write some DSP code to do it. Helloooo.... Jim Fosgate built his surround processor using tubes and Dolby ported it to DSP when he sold it to them. I really think the guy is a bit too impressed with himself and a bit clueless about the real world.

[augerpro]

I just received my 10" waveguides from Earl. They were cut off from the Nathan 10 baffle, which I expected. And it makes them easier to integrate into my baffle than the fiberglass ones I've seen.










They are going to take some finishing but I think they're usable. The plastic is pretty thick. One is smooth and the other (redder) one has some sanding marks that will have to smoothed. The rear on both is rough and will require some bondo and sanding since this will be on an open baffle. The only thing that bugs me is the plexiglass mounts (which are nicely machined) don't fit the back of the waveguide well. The center hole is supposed to match the angle and size of the waveguide throat and it's actually a bit larger, by about 1mm. Pretty big error considering all that Earl has made about it. If it was just smaller I could bondo it inside the waveguide and smooth it out. The only way I see to make this work is to mill down the entire back until the waveguide throat is the same size, kinda like you have to do with the MCM waveguide. I'll have to route it by building a jig using the baffle as a base and reference and hope the waveguide is not mounted even slightly off kilter or I'll end with an oblong throat.

[EdL]

Josh,
Thanks for the post with photo and explanation.. I couldn't imagine what the plexiglas was for.

[JoshK]

It took me a second to figure out what the plexiglas was for when I first opened up the box, then I started to think how the CD's were suppose to mount to the particle board and then the light went off.

Brandon,
That sucks about the plexiglas not matching the diameter of the waveguide. That would upset me. Its going to be a chore to fix. Mine match pretty well actually, I guess I am lucky.

[JoshK]

Has anyone inquired whether Assistance Audio has any newer DDS ENG 1-90's? I wonder if they are shipping good ones yet. If so, that might be a better option for those wanting 10" waveguides. Even at ~$100/ea, they are cheaper.

[Saurav]

After seeing what people are going through with these, I'm feeling pretty good about my XT1086's

[ThomasW]

Has anyone inquired whether Assistance Audio has any newer DDS ENG 1-90's? I wonder if they are shipping good ones yet. If so, that might be a better option for those wanting 10" waveguides. Even at ~$100/ea, they are cheaper.

That question will be answered shortly. I've been trying to get a hold of Jack to place an order. So far all I've gotten is a "inbox is full" message from his answering machine. He maybe off doing a gig so I'll keep trying.

[Dennis H]

John Janowitz let this tidbit slip over at AVS. "There don't seem to be too many good waveguides that are readily available though so we started looking at what we could do to solve that issue as well." If he pairs up with Earl Geddes to CNC his boxes, maybe they can cut some kind of deal so John can CNC horns as well?

[Mark Seaton]

John Janowitz let this tidbit slip over at AVS. "There don't seem to be too many good waveguides that are readily available though so we started looking at what we could do to solve that issue as well." If he pairs up with Earl Geddes to CNC his boxes, maybe they can cut some kind of deal so John can CNC horns as well?

John is helping Earl out by supplying the flat kits for the kits Earl is offering, but Earl isn't involved with what John was referring to. :roll:

[CraigJ]

Any updates on how these projects are coming?

Thanks,

Craig

[Quwiksilver]

We already know the profile, but I don't know CAD.

I do. What's the profile? I'll model something up.

[Quwiksilver]

I do. What's the profile? I'll model something up.

Nevermind. I found it.

Here's a 10" waveguide with a 1" throat diameter, theta = 60°, 0.06" cone offset, 0.0185 back offset. The waveguide is 0.25" thick with a 0.75" rim to easily countersink into to 3/4" MDF. You'd need Qty 8 M5 X 30mm Socket Head Cap Screws with accompanying hurricane or t-nuts.

There's a front and rear view, as well as closeups of the throat with and without a mini SS 6600 mounted in it.

I can't believe I'll actually get PAID to do work like this someday (hopefully). Better get back to studying.

[Quwiksilver]

Just peeked at McMaster-Carr. Looks like they sell some 10" diameter UHMW bar stock (Part #8701K721) that goes for $237.25/ft. Given the thickness of the waveguide, you could make 4 waveguides/ft, or about $60 per waveguide in material costs.

I imagine the model above would also have to include features to be able to easily bolt on the driver. I'm not sure what the driver looks like or the best way to do that.

Who's got a CNC? :T

[Arc00]

If you have a CAD file, Elemental Designs might be able to do it for you. But it will have to be from MDF.

What if you made two, then used a epoxy or something and just make 1 mold for multiple WG's?

[Quwiksilver]

Version 2.0

Now we're getting somewhere. I've added a compression driver and devised a better method of mounting it. It is similar to the way Earl does it except that it contains a clearance hole large enough to allow the driver itself to pressed against the waveguide. This keeps the waveguide profile which the driver sees smooth and in one piece. The driver's mounting plate (0.25" aluminum) is then bolted to the waveguide with a small clearance (0.025") such that these bolts compress the faceplate of the driver against the throat of the waveguide. This should be a pretty stiff arrangement. I also eliminated the profile on the backside of the waveguide as it was unnecessary.

Here's a question: would it be better to mount the waveguide to the enclosure using fasteners from the backside to keep the "perfect" shape of the waveguide uninterrupted?

Scott

[Xander]

Solidworks, right? I can tell because of the cosmetic thread markings on the ends of the bolts showing through the other materials. They bug me, so I always hide them, haha.

If I knew more about the drivers I could help out too, but it looks like you have it under control.

[Quwiksilver]

Solidworks, right? I can tell because of the cosmetic thread markings on the ends of the bolts showing through the other materials. They bug me, so I always hide them, haha.

If I knew more about the drivers I could help out too, but it looks like you have it under control.

Heh...yep. I don't really see anything "cosmetic" about them. They're ugly.

This model can be adjusted for different theta angles, waveguide diameter, throat diameter, etc with just a few minutes work. Unfortunately, Solidworks 2008 and prior (I'm using '07) can't generate curves directly from an equation. Instead of plugging in the simple oblate spheroid equation I have go and generate a set of points on excel to define the curve. The curve can then be swept into the solid circular shape. I'd have preferred a direct method, but with 10,000 data points, accuracy is more then enough. I started with catapult's excel spreadsheet from this thread over at diyAudio.

Next we need to import a few waveguides into ANSYS to see how they behave. After finals maybe.

[augerpro]

Cool stuff Q. Who's formula were you using? I have any idea adapting Joshk's spreadsheet for use with a regular dome tweeter. It would be great to get into a form that I could hand to some CNC operator and do some waveguide prototypes.

[Dennis H]

I think that's one of my spreadsheets. You can download it from the DIYaudio thread (I'm catapult over there). I dunno if Josh ever modified his to allow a non-zero throat angle. Read the thread for all the dirty details. The one thing mine doesn't do is round over the horn at the baffle surface and I think Josh's does that so pick your poison.

[augerpro]

I have Josh's and yours. Funny thing is when I went looking for Josh's the one I thought it was had Ed as the owner. But it is indeed the one as it allows for the roundover. That's why asked the question actually. Josh does have an entry for the driver exit angle but for some reason it doesn't work for me. How hard would it be to implement the roundover calculation Dennis? That's pretty key to the design, particularly if I try to adapt it to a dome tweeter for a typical size baffle. John K's spreadsheet also doesn't allow for the roundover.

Also if I were using say theta=45 degrees, and wanted it to just stay a straight cone, how would I do that? I tried used the goal seek on F9 as directed, but inputting 45 degrees there gives a screwed up calculation.

And what does the cone offset figure do?

[EdL]

OS spreadsheet

I "may" have the spreadsheet you want. It does allow entry of the throat angle and rounding of the waveguide to the baffle.

I originated the spreadsheet that Josh improved greatly with his abilities in excel. It would not be in the state it is without his help.

I have no way of providing a link to the current worksheet. I'll e-mail a copy to all who request. I have no objections should someone want to post it with access through a link.



Geddes has reviewed it and not stated any objections...probably as good an endorsement you'll get...

[augerpro]

That's the one Ed. One problem: changing "throat radius of driver" and "Driver's plug exit angle" don't seem to do anything for me.

[EdL]

There was a worksheet which had the best of intent. It included some incomplete links. The fix was discussed @ diyAudio. Send me an e-mail, I'll send you the latest worksheet.

[augerpro]

PM sent

[CraigJ]

A Quote from Earl regarding his Abbey/Nathan/Summa, as of 12/18: "I'm still looking for that "sweet spot" of the design, maybe a 12" woofer and 15" waveguide, maybe a 12" woofer and a 15" x 10" elliptical, I don't have enough data yet to know. But of the data I have the Abbey is the clear leader in terms of best compromise." Further reading: https://www.diyaudio.com/forums/show...6#post1689396I didn't think I'd hear him think out loud like this, guess I'll wait on a kit.....

[goskers]

I have just finished a set of Nathan's a little over a week ago. As of last night I have also completed the multiple sub arrangement that Geddes recommends.
It has been requested that I post all of the build documents here at HTG. I will probably do so tomorrow if the music doesn't get in the way :T

Being able to hear the Geddes designs in his own place a few times had me a little nervous about the speakers in a non-optimized room. I can place all of my fear on the back burner now as they do all that I wanted them to do. :E

New thread to follow...

[penngray]

Heh...yep. I don't really see anything "cosmetic" about them. They're ugly.

This model can be adjusted for different theta angles, waveguide diameter, throat diameter, etc with just a few minutes work. Unfortunately, Solidworks 2008 and prior (I'm using '07) can't generate curves directly from an equation. Instead of plugging in the simple oblate spheroid equation I have go and generate a set of points on excel to define the curve. The curve can then be swept into the solid circular shape. I'd have preferred a direct method, but with 10,000 data points, accuracy is more then enough. I started with catapult's excel spreadsheet from this thread over at diyAudio.

Next we need to import a few waveguides into ANSYS to see how they behave. After finals maybe.

What happened to this project????

I really think someone should be able to create OS waveguides for much less then Geddes charges.

[jdc0589]

If you have a CAD file, Elemental Designs might be able to do it for you. But it will have to be from MDF.

What if you made two, then used a epoxy or something and just make 1 mold for multiple WG's?

If you can get a good wooden piece made it would be very easy to make a NICE fiberglass mold to produce fiberglass plugs (unlike the terrible picture posted earlier. Damn, im a composite novice and could have done a lot better than that.)

[penngray]

If you can get a good wooden piece made it would be very easy to make a NICE fiberglass mold to produce fiberglass plugs (unlike the terrible picture posted earlier. Damn, im a composite novice and could have done a lot better than that.)

There are injection molding companies that should be able to do this stuff no??

I just seems so easy considering we have all the data and once a template is made.

There is a reason JBL, QSC, etc sell waveguides/horns for $10 and Geddes sells his for $400. There is no secrets to his waveguid either it just amazes me someone has not mass produce it (Ie....JBL or others.)

[NEO Dan]

Penn,
It's all about the $$$ for tooling.
Figure on two or three CNC protos from wood costing who knows how much.
And then you get down to the actual tooling for production and pricing per piece.
While you are at it you would want to get some CD's OEM'd for your WG so you can ensure results and sell it as a package at a moderate profit without gouging the customer.
If you do the math and it seems reasonable to you to invest 10-15K in a product that should sell for ~$150 a pair CD's and WG's then let's start the ball rolling. I have a friend that has injection molds for plastics and also metal castings tooled in Asia regularly. I can get a quotation of free. BMS would be my first choice on the custom CD's as they offer quality without the pretentious pricing.

[penngray]

Thanks Dan,

If I was an engineer at JBL or QSC and I followed any of the DIY forums I would pick up on all the discussion wrt Geddes Waveguides then just build a similar OS or eliptical shape myself. They seem to have many different horns/waveguides that cost nothing so they have the ability to make a new waveguide.

I can only think of a couple reasons this isnt happened.

1. They are not following DIY projects and great new discoveries.

2. They do not care about all this HOM/Horn sound because their business caters to a market that is not impacted by those preceived problems.

[Dennis H]

JBL knows all about OS waveguides as they worked with Earl to try to patent them back in the day. They even advertise them in their consumer line.

http://www.jbl.com/home/products/product_detail.aspx?prod=ES90BK&Language=ENG&Country=us&Region=USA&cat=BFS&ser=ES

"The tweeter is mounted in our proprietary Elliptical Oblate Spheroidal™ (EOS) waveguide for improved high-frequency dispersion anywhere in the listening room."

[JonMarsh]

JBL knows all about OS waveguides as they worked with Earl to try to patent them back in the day. They even advertise them in their consumer line.

http://www.jbl.com/home/products/product_detail.aspx?prod=ES90BK&Language=ENG&Country=us&Region=USA&cat=BFS&ser=ES

"The tweeter is mounted in our proprietary Elliptical Oblate Spheroidal™ (EOS) waveguide for improved high-frequency dispersion anywhere in the listening room."

Any guesses for how low that device maintains pattern control? :W the first two don't count. :B